> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://nvidia-cuvs.docs.buildwithfern.com/cuvs/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://nvidia-cuvs.docs.buildwithfern.com/cuvs/_mcp/server.

# User Guide

Use these guides when you are ready to apply NVIDIA cuVS APIs, benchmark algorithms, or integrate NVIDIA cuVS into a larger product.

## Guides

- [API Guide](/user-guide/api-guides): find task-focused NVIDIA cuVS API examples for clustering, vector indexing, preprocessing, common types, and supporting routines.
- [Benchmarking Guide](/user-guide/benchmarking-guide): learn how to compare vector indexes fairly and use cuVS Bench for reproducible benchmark workflows.
- [Field Guide](/user-guide/field-guide): find practical material for compatibility, integration patterns, runtime compilation, and extension points.
- [References](/cuvs/user-guide/references): cite the research papers behind NVIDIA cuVS vector search, preprocessing, clustering, and GPU primitives.